News archive
Recruiting, HR & AI news: November 14, 2026
Saturday, November 14, 2026. 26 stories, each linked to its source.
Ranked by importance within each day. Pick a topic to see it newest first.
Saturday, November 14
Get the news by email
Free newsletter
Get the Work Insiders weekly
One email a week: the recruiting and HR tech news that mattered, the tools we just added to the directory, and one practical guide. No spam, and you can unsubscribe in one click.
A funding announcement is a buying signal.
Scaling cold email means buying mailboxes in bulk.
The same 1,200 CVs a year cost $84 at $0.07 a CV and $3,588 inside an ATS plan.
Google has rolled out Dynamic Workflows in Antigravity, its agentic coding app, letting the main agent write a short Python script that runs other agents in parallel or in sequence instead of handling a big task in one long conversation. Users start one with /run-workflow or by asking in plain language, review the script before it runs, and can save it to re-run the same flow later. Each subagent works in its own conversation and returns a structured result checked against a JSON Schema, which keeps logs and file reads out of the main chat. The feature is available on Google AI Pro and Ultra plans in the Antigravity app and CLI.
Sierra has released version 1.0 of Personal Agent Protocol, or Poppy, an open standard for how personal AI agents find and act on companies' websites, APIs and customer-service agents on a user's behalf. The protocol builds on existing standards such as MCP and OpenAPI and uses OAuth, so customers control what their agent can and can't do with each account. Shopify has enabled Poppy for every merchant on its platform, Meta's Muse agent now looks for it first on every site it visits, and Sierra says more than 100 companies, including Salesforce, Zoom, Twilio, Delta and Intuit, have joined as design partners. Sierra still labels the spec a draft that may change in ways that break existing setups.
Voice AI company Bland has introduced Tofu-1, a full-duplex speech-to-speech model that listens while it speaks, so callers can interrupt or correct it mid-sentence the way they would with a person. Instead of converting speech to text and back, it works on the audio itself, which Bland says lets it tell speakers apart, separate voices from background noise and keep the call moving while it looks things up. Bland is pitching it at banks, health systems and government agencies, where calls often come from cars, hospital rooms or weak signals. Tofu-1 is in a research preview open to a limited number of enterprises.
World Labs Opens Atlas to Everyone, Letting Anyone Turn Photos or Sketches Into Explorable 3D Worlds
Fei-Fei Li's World Labs has opened Atlas, its world model, to anyone who signs in, ending partner-only access that began when the model was announced in September. Two browser apps run on it: Viewfinder rebuilds a scene from photos or a single image and lets users move a camera through it, while Chisel lets users block out a layout with simple shapes and see every view update. Results export as keyframes, video clips or 3D Gaussian splats, with tools to convert splats into meshes for games and to relight scenes, plus a developer API that works with tools such as Blender. World Labs has not published a paper or model card, and its quality figures remain company-reported.
No stories on this topic yet.



